Synthetic monitoring

The Synthetic API monitoring workflow runs hourly and can be started manually for staging, production, or both. Configure repository variables:

STAGING_API_BASE_URL=https://staging.example.com
PRODUCTION_API_BASE_URL=https://api.example.com
SYNTHETIC_PROFILE=full

The monitor checks /health/live, /health/ready, /metrics, and the full profile compatibility endpoint /api/v1/health. Minimal-profile projects skip the compatibility endpoint when profile=minimal is selected. Health requests must expose the correlation X-Trace-Id header, and metrics must include the starter request counter.

Failures create one deduplicated [synthetic-monitoring] issue per environment. Use the incident response runbook to triage the issue; do not put credentials, tokens, request bodies, or personal data in URLs or issue descriptions.

Dependency probes

observability/prometheus/aspnet-starter-dependency-alerts.yml contains optional rules for blackbox or platform probes labeled with environment, service, and dependency. Import only the rules for dependencies actually used by the generated project. Redis, email, and messaging alerts are warnings because those adapters are optional and may be configured fail-open; PostgreSQL readiness is critical for the full profile.